Man-Made Horrors: August 2024

Man-Made Horrors for Your Pleasure!

Welcome to our second monthly newsletter, wait does this mean we’ve formed a habit? We’ve kept a schedule? Inconceivable! Though this word might not mean what I think it means, we’re pretty stoked with how many of you have joined us on this ride. Also, if you’ve missed our most recent video where we explore a miniaturization technique called microdots which was used extensively by spies world wide, you can catch it here.

In June, some of you may have attended Open Sauce where Justin participated in the Mad Science panel, with Tom (Explosions&Fire), Nigel (NileRed) and Cody (Cody’s Lab) moderated by Kevin (TheBackyardScientist) where the lot explored how they’d destroy the world for entertainment, or not. The event is a classic for us now, and we’ll be returning next year for more creator insanity. If you want to meet some of your favourite science and maker youtubers, it’s the place to be and tickets are already on sale for next year!

Other than being invited we are not associated with the event, but we believe the event sells itself, and love attending to meet the fans of the channel who show up.

From the Cutting Room Floor

In the latest neuron video we covered a variety of arrays that we had fabricated to test different electrode arrangements and the effects of our manufacturing process would have on their functionality. The gold covered copper trace piercing out of a coverlay allows us to have relatively small electrode heads, yet they’re not quite as small as the plasma deposited arrays made by Multichannel Systems. So we were afraid we would require multiple electrodes receiving the same pulsed signal to have confirmation of a positive firing. This led us to come up with the trigger pad concept with a high neuron seeding density, where a hole in the coverlay has multiple electrodes close to each other controlling 1 action. With a high neuron density then the majority of these electrodes receiving a signal could constitute as our desired output. Interestingly enough, through our experiments it would appear that even with a medium to low seeding density of neurons, a single electrode within a pad can reliably pick up the neurons firing. This wasn’t the result we were expecting, our homemade electrodes arrays are picking up the voltage spikes from a single neuron! We’ll definitely take that as a victory! Our electrode density is obviously lower than the traditional commercial arrays, but because we’re looking for a single output signal from the neurons, we don’t need to track the signal’s propagation across the entire dish. Something that would be more relevant were we studying the neurons themselves and their transmission mechanisms, but we just need one actionable output signal per game action/interaction. For all intents and purposes, we are going to be using the neurons as a neural network, and so fundamentally the exact internal mechanisms of its decision making process can remain a black box for it to play a video game.

The design of the arrays includes a feedback mechanism that is intended to provide a confirmation signal to the neurons when they have actually accomplished a task. This will of course need to be tested, as we are not yet sure if the simple disappearance of an input signal following a neuron’s firing would be enough of a reward for the action. Therefore we built in the possibility to test either method. As for the types of signals, we know that neurons seem to “prefer” smooth regular waveforms and are disrupted by irregular noise, such irregular noises can be used as a negative reinforcement tool. At least that’s the plan.

But, what about the neurons themselves, what do they feel? Well that’s a bit more complicated. We named the machine the Torment Nexus, mostly because the thought of a dish of neurons only ever experiencing playing doom is so fitting. But are the neurons tormented? Are they in pain? For that matter, what is pain? What about suffering? Well, we have to look at this under a philosophical lens, and Chalmers in his Philosophy of Mind describes an Easy and a Hard problem to consciousness. The Easy problem is self evident, what happens when we feel pain? A stimulus activates a receptor which sends a signal up through the nervous system declaring that there’s a situation that needs a reaction to be addressed. Pain is a motivator, a call to action, hell even boredom triggers a sensation neurologists would describe as painful. Thus we can answer the Easy question mechanistically, but fundamentally this does not ascribe any value judgment on the feeling of said pain. The Hard question therefore is about the fundamental experience of a feeling, in this case pain; we’ve all stubbed our toe and cried out in pain when there wasn’t any pain, we let out a yelp that was only a contact alarm. But it felt like pain in the moment, it was expressed as pain, without any damages to the body. Our higher consciousness can experience pain when there isn’t any or pain when there aren’t necessarily external stimuli that would cause it. 

In the same sense, we the humans watching the dish of neurons stuck playing doom for its whole existence can ascribe to it the experience of being tormented for all eternity. Yet the same control scheme could be used to control a drone. For example the Asteroids playing array we designed could be used to control the heading of a drone in flight, with the targets being the waypoints and altitude. To the dish of neurons, there would be little to no appreciable change in the input signals. The dish lacks the higher consciousness to ascribe meaning and context to any of its experiences. If somehow our neuron plates start to develop a contextual understanding of their situation we’d have a world first and we’d be trying to get confirmation studies going ASAP. 

In The Works

As we mentioned in our Connecting Living Neurons to a Computer video, Multichannel Systems sent us a variety of Multi-Electrode Arrays (MEA) and their MEA2100-Mini recording system to test our ability of getting signals out of neurons. With this we validated the design of our PCB style arrays and made a huge leap forward on the project. But seeing as we have a really cool toy in the lab for the whole summer, we’ve been testing it out by listening to Jerry and other organisms. For anyone who isn’t in the know, Jerry is the lab’s pet slime mold (scientific name: Fuligo septica) and we’ve put Jerry through a variety of mazes in the past which it surprisingly solved.

We're also having some fun with optics and Justin's new/old film camera making old spy tech, holograms and photon sieves. The next video from these experiments will be out shortly, and as usual we'll send out a notice as soon as it's published! Honestly a laser and optics project that wasn't a complete pain to pull off is quite the novel experience for us.

The neuron array poster production has been moving forward and we will start shipping them out in a couple weeks. If you want one, they’re still available in our shop from this production run.

Neat Finds

Using powder metallurgy to make knives?

NOAA discovered the largest deep sea Coral Reef?

Modeling neurons, where does one start?